Cost-Benefit Analysis: Six Case Studies (With Implications) and An Essay on Regulatory Management

The following post comes to us from John C. Coates IV, John F. Cogan Jr. Professor of Law and Economics at Harvard Law School.  It is based on two of his recent articles, “Cost-Benefit Analysis of Financial Regulation: Case Studies and Implications,” which is forthcoming in the Yale Law Journal and is available here, and “Towards Better Cost-Benefit Analysis: An Essay on Regulatory Management,” which is forthcoming in Law and Contemporary Problems and is available here.

The 2010 Dodd-Frank Act mandated over 200 new rules, bringing renewed attention to the use of cost-benefit analysis (CBA) in financial regulation. CBA proponents and industry advocates have criticized the independent financial regulatory agencies for failing to base their new rules on CBA, and many have sought to mandate judicial review of quantified CBA (examples of “white papers” advocating CBA of financial regulation can be found here and here). An increasing number of judicial challenges to financial regulations have been brought in the D.C. Circuit under existing law, many successful, and bills have been introduced in Congress to mandate CBA of financial regulation.

In two related papers, I take up the topic of CBA of financial regulation. In the first paper, “Cost-Benefit Analysis of Financial Regulation: Case Studies and Implications,” I develop detailed case studies that collectively show that precise, reliable, quantified CBA is simply not feasible in a representative sample of six significant financial regulations. Instead, efforts to quantify CBA routinely require multiple subjective, contestable judgments, producing what can only be called “guesstimates”—extremely wide ranges of equally defensible net costs and benefits—so wide that the effort cannot meaningfully “guide” assessments of the rules. In the second paper, “Towards Better Cost-Benefit Analysis: An Essay on Regulatory Management,” I discuss possible ways to improve CBA as applied to financial regulation and argue that the best methods will not include simple legal mandates, as in proposed bills, or stringent judicial review, as many have advocated. Instead, the better approach, one more likely to produce reliable CBA and improve regulatory outcomes, involves drawing on a wide range of techniques within the “management” toolkit, including funding, governance, disclosure, regulatory design, and agency culture.

The paper first clarifies the meaning of CBA. CBA is often used to mean policy analysis, but also to mean types of laws mandating the use of CBA or providing for judicial review of CBA or its inputs. CBA is also sometimes used to describe a general conceptual framework for evaluating regulations, roughly amounting to applied welfare economics, which is a sensible framework for general policy analysis, but it is also sometimes expected to produce specific quantified estimates of the effects of a rule—“quantified CBA.” Finally, CBA is often viewed as a clearly good thing as a policy matter—enhancing transparency and democracy, and disciplining regulators—but the history of CBA suggests that it can also produce a darker mix of effects, including rent-seeking, increases in regulatory discretion, and reductions in transparency. The paper then reviews current law of CBA relevant to financial agencies, and critiques the recent spate of D.C. Circuit decisions striking down regulations on the supposed ground of inadequate CBA.

The core of the paper is a close examination of what efforts to quantify CBA might produce, or in two cases what such efforts did produce, as applied to six representative and major financial regulations:

  1. The SEC’s disclosure rules under Sarbanes-Oxley Section 404,
  2. The SEC’s aborted mutual fund governance reforms,
  3. Basel III’s heightened capital and liquidity requirements for banks,
  4. The Volcker rule,
  5. The SEC’s cross-border swap proposals, and
  6. The FSA’s mortgage market reforms.

In the case studies, the paper attempts to advance the substantive project of quantitative CBA of financial regulation itself, by specifying in more detail than prior research how CBA could be applied in specific contexts, and by more carefully describing, clarifying, and specifying sources of sensitivity of the results of quantified CBA in each study. The primary general conclusion of the case studies is that CBA of such rules cannot be reliably and precisely quantified with current research methods.

Rather, CBA for major financial regulations entails (a) causal inferences that are unreliable under standard regulatory conditions; (b) using problematic data; and/or (c) the same contestable, assumption-sensitive macroeconomic and/or political modeling used to make monetary policy, which even CBA advocates would exempt from CBA law. Because finance is central to the economy, because it is social and political, and because it is non-stationary, regulation of finance and financial markets is different in kind from many other kinds of regulation, which focus on activities that can be more reliably studied. As a result, judicial review of quantified CBA in the financial context is more likely to camouflage discretionary choices than to discipline agencies or promote democracy.

Nevertheless, CBA is a useful conceptual framework for financial regulation, and quantified CBA is a worthy long-term research goal. CBA should be conducted only to the extent the relevant financial agencies choose to do so. But the agencies could be better equipped and funded to conduct CBA, and their governance could be improved to make it possible to do so.  Laws hampering effective CBA should be modified, such as those imposing process delays on one-time efforts to collect CBA-relevant information or barriers to regulatory experiments that can generate much more reliable information about the effects of regulation.